Political Decision Making: Diversity or Systematic Error?


sysquality_sys_randomI have ignored group decision making to a large extent, but bootstrapping has somehow brought me back to it–especially dialectical bootstrapping which seems to be one person group decision making. Obviously, group decision making is important.  This post will focus on political decision making. Two books from 2007, Scott Page’s: The Difference — How the Power of Diversity Creates Better Groups, Firms, Schools and Societies and Bryan Caplan’s: The Myth of the Rational Voter–Why Democracies Choose Bad Policies look at it from far apart.

Caplan makes the argument that the public is systematically stupid about economic issues and this eventually translates into voting stupidly which becomes the government doing stupid things.  According to Caplan, the general public is systematically stupid because of what he calls rational irrationality.  He says that they have preferences over beliefs. By this he means that certain beliefs are preferred for reasons other than their truth value. This could be that the belief fits with the image that they have for themselves and fits in with other beliefs that they hold. Now, this was apparently news to Caplan. (See posts Bidirectional Reasoning or Justifying our Decisions: Great for Plausible Deniability, not so Great for Medical Diagnosis  for examples.) I would insist that economists do the same thing. Caplan more or less says that our government would produce better policies if we limited voting to smart educated people or gave their votes more weight.

Caplan’s best point is that a voter is different than a consumer. This is largely because the chance for any individual to change the result of any election is very small so that you can vote your conscience, vote your beliefs, vote your worldview, without any real chance of consequences to your life. Now this seems more true at a national level than in the school board election. I also note that we have a representative democracy and that the founders were aware of many of our human foibles. Cass Sunstein (post Going to Extremes and Why Society Needs Dissent ) points to the American Constitution as an attempt to create a deliberative democracy that combines accountability with a measure of reflection and reason giving. Some wanted the Bill of Rights to include a “right to instruct” which would bind representatives to vote with the citizens. Sunstein points out that today, the majority might want to include the “right to instruct”.  Sunstein quotes Roger Sherman for the reasons that this is not a good idea:

…I think, when the people have chosen a representative, it is his duty to meet others from the different parts of the Union, and consult, and agree with them on such acts as are for the general benefit of the whole community.  If they were to be guided by instructions, there would be no use for deliberations.

cartoon2So I question that economists are better voters than the general public, and also question that completely free trade is always a good thing for everyone which Caplan seems to believe.  The law of comparative advantage is a cool thing–a lot like the miracle of aggregation. It works ceteris paribus. In my opinion, Caplan needed to make the book sell and for important readers to take notice. He did it well. He oversold a bit, but he provides many caveats. Democracies can make stupid policies and voters do vote for Warren Harding. We do need to do better, and markets could help sometimes.

Scott Page does not seem to need to let you know that he is the smartest guy in the room as much as Caplan. Scott Page has formalized what Caplan calls the miracle of aggregation or what Suroweicki calls the wisdom of crowds into the diversity prediction theorem which simply states that the crowd’s error = avg error- diversity.  This means that the more diverse the crowd the smaller its error. Building on this Page and his colleague, Lu Hong have also created a more sophisticated model of agents who possess predictive models. This second approach builds upon and extends traditional statistical approaches to characterizing collective wisdom by demonstrating how collective accuracy requires either individual sophistication/expertise or collective diversity.

According to this model the prediction of a crowd of people can be thought of as an average of the forecasts produced by models contained within the individual’s heads. Thus, collective wisdom depends on characteristics of the models people carry around in their heads. For collective wisdom to emerge those models must be sophisticated, or they must be diverse. Ideally, crowds will possess both. A collection of people becomes likely to make a bad choice if they rely on similar models. This idea aligns with the argument made by Caplan that people make systematic mistakes. Page and Hong note that if everyone leaves out some relevant feature of the world in constructing their models, then the crowd cannot be accurate. But here they depart from Caplan and state that the cause of wise or mad crowds is not just the intelligence of the people who make up that crowd. They insist that enough collective diversity can make up for lack of individual sophistication. This presents an interesting challenge for democracy. Not only should democratic institutions encourage sophisticated thinking, they should also support diversity. According to the diversity prediction theorem, in the aggregate, collective diversity matters as much as individual sophistication. Page and Hong add that diversity might be easier to attain.

Peter Schuck in his book Why Government Fails So Often acknowledges several kinds of irrationalities that impact politics.  First, he looks at the work of Kahneman and Tversky that illuminated the multiple types of bias to which we are subject (post Prospect Theory). Second, he points to Sunstein who is concerned with group polarization and cascades, and who proposes “nudging” the public toward more rational decisions. Third, he examines “cultural cognition” led by Dan Kahan (post Numeracy and Cultural Cognition). Cultural cognition is the propensity of people to assess objective evidence in ways that try to maintain consistency with their existing ideological identities such as individualism/hierarchy or communitarianism/ egalitarianism.  The fourth body of research is by Jonathan Haidt (see above) that examines our modular brains. Schuck also looks at Caplan’s work and gives it considerable credit, although Schuck offers possible remedies.

Caplan, B.(2007). The Myth of the Rational Voter Why Democracies Choose Bad Policies. Princeton University Press: Princeton and Oxford.

Hong, L. & Page, S.(2012). “Some Micro-foundations of Collective Wisdom.” in Collective Wisdom: Principles and Mechanisms edited by Hélène Landemore, Jon Elste, Cambridge University Press: Cambridge.

Page, S. (2007). The Difference How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies. Princeton University Press: Princeton and Oxford.

Schuck, P.(2014) Why Government Fails So Often and How It Can Do Better. Princeton University Press: Princeton and Oxford.

1 thought on “Political Decision Making: Diversity or Systematic Error?

  1. Pingback: Nervous States: Counterpoint | Judgment and Decision Making

Comments are closed.